Predicting the potential distribution of the endemic snake Spalerosophis microlepis (Serpentes: Colubridae), in the Zagros Mountains, western Iran

2017·1.2 MB·English

by  Hosseinzadeh, M S

SALAMANDRA 53(2) 294–298 15 May 2017 ICSoSrNre 0sp03o6n–d3e3n7c5e
Correspondence
Predicting the potential distribution of the endemic snake
Spalerosophis microlepis (Serpentes: Colubridae),
in the Zagros Mountains, western Iran
Mahboubeh Sadat Hosseinzadeh1, Parviz Ghezellou2 & Seyed Mahdi Kazemi3,4
1) Department of Biology, Faculty of science, Ferdowsi University of Mashhad, Mashhad, Iran
2) Department of Phytochemistry, Medicinal Plants and Drugs Research Institute, Shahid Beheshti University, G.C. Evin, Tehran,
P.O. Box 19835-389, Iran
3) Department of Biology, College of Sciences, Qom Branch, Islamic Azad University, Qom, Iran
4) Zagros Herpetological Institute, 37156-88415, P. O. No 12, Somayyeh 14 Avenue, Qom, Iran
Corresponding author: Seyed Mahdi Kazemi, e-mail: [email protected]
Manuscript received: 24 January 2015
Accepted: 5 January 2016 by Andreas Schmitz
Species distribution models (SDM) are geographical mod- rouz 2005, Baig & Masroor 2008, Rastegar-Pouyani et
els of biospatial patterns in association with environmental al. 2008, Schätti et al. 2009, Schätti et al. 2010). The lat-
factors (Franklin 1995). The predictive models of species’ ter species is distinguished from Spalero sophis diadema by
geographic distributions are important for a variety of appli- distinctive morphological characters including 41–45 mid-
cations in ecology and conservation (Graham et al. 2004). body scale rows (Marx 1959, Schätti et al. 2009; Fig. 1).
Therefore, species’ distributions can be modelled to provide However, Spalerosophis microlepis is a rare and poor-
suitable information for many rare and poorly known taxa. ly known species and its range is uncertain (Marx 1959,
Some snake species are particularly difficult to detect due Gholamifard 2011). It occurs in western and central Iran
to their low densities, elusiveness, or long periods of inac- as well as the Zagros Mountains, in Ilam, Lorestan, Fars,
tivity (Seigel 1993). Thus, their distribution ranges may be Khuzestan, Hamadan, Markazi, Qom, Kerman, Chahar
underestimated and less well known than in other reptiles Mahall-va-Bakhtiyari, Kohkiluyeh-va-Boyer Ahmad, and
(Santos et al. 2006, Bombi et al. 2009). SDMs can be used Isfahan provinces (Latifi 2000, Rastegar-Pouyani et
to fill these knowledge gaps by mapping potential distribu- al. 2008, Gholamifard 2011, Moradi et al. 2013, Kaze-
tion ranges and so identify sites at which searches are more mi et al. 2015). Records of this species from Semnan, west-
promising than at others and should be considered for con- ern Yazd and northern Hormozgan need to be confirmed
servation programmes (Peterson et al. 2000). (Baig & Masroor 2008, Schätti et al. 2009). Addition-
The genus Spalerosophis Jan, 1865 (type species Spalero­ ally, the species might be present in Iraq, although this
sophis microlepis) includes six species, S. arenarius (Bou- requires confirmation (Firouz 2005, Baig & Masroor
lenger, 1890), S. atriceps (Fischer, 1885), S. diadema 2008, Schätti et al. 2009, Gholamifard 2011). Accord-
(Schle gel, 1837), S. dolichospilus (Werner, 1923), S. joseph­ ing to Latifi (2000), S. microlepis has been reported to oc-
scorteccii Lanza, 1964, and S. microlepis Jan, 1865 (Sindaco cur in mountainous areas, foothills, fields, grasslands, and
et al. 2013, Uetz 2015). The genus occurs in arid and sem- semi-desert regions.
iarid regions, the Saharo-Sindian region, from North Af- The aims of this study are to provide a comprehensive
rica in the west through Arabia, Iran, Pakistan, to central distribution map of S. microlepis, to confirm the presence
India in the east (Baig & Masroor 2008, Sindaco et al. of S. microlepis in doubtful localities, and to identify the
2013, Uetz 2015). Four taxa of two species of the genus have environmental variables associated with the predicted dis-
been recorded from Iran, i.e., Spalerosophis diadema cliffor­ tribution of S. microlepis using a maximum entropy distri-
dii (Schlegel, 1837), Spalerosophis d. diadema (Schlegel, bution modelling approach.
1837), Spalero sophis d. schirazianus (Jan, 1863) and Spale­ All records of S. microlepis are based on our own field-
rosophis micro lepis Jan, 1865 (Marx 1959, Latifi 2000, Fi- work as well as those from the literature (Frynta et al. 1997,
© 2017 Deutsche Gesellschaft für Herpetologie und Terrarienkunde e.V. (DGHT), Mannheim, Germany
A2v9a4ilable online at http://www.salamandra-journal.com
Correspondence
Table 1. Percentages of contributions of variables included in the best-fitting distribution model for Spalerosophis microlepis.
Environmental variables Percent contribution Permutation importance
Bio18, precipitation in the coldest quarter 40 38.1
Bio12, annual precipitation 20 4.4
Bio8, mean temperature in the wettest quarter 15.6 0.4
Slope 10.5 23.6
Bio17, precipitation in the driest quarter 8.5 0.8
Bio2, mean diel temperature range (monthly mean [max.–min.]) 3.1 0.4
Bio14, precipitation in the driest month 1.1 0.3
Bio5, maximum temperature in the warmest month 0.6 16.4
Bio7, annual temperature range 0.5 15.6
Bio15, precipitation seasonality <0.1% <0.1%
Latifi 2000, Schätti et al. 2009, Sindaco et al. 2013). Ad- from species distribution modelling (Rissler et al. 2006).
ditionally, we included point localities based on museum Then, 10 out of 21 environmental variables were chosen and
specimens. The records that were used in this study are from used in this study; see Table 1 for more details.
the following museums: The Natural History Museum, Lon- Maxent is a modeller approach associated only with
don (BMNH); Field Museum of Natural History, Chicago species presence data that enables the construction of well-
(FMNH); Muséum d’Histoire naturelle, Genève (MHNG); fitted predictive performance and ecological data. It is con-
National Museum of Natural History, Washington D.C. sidered one of the most efficient approaches for predicting
(USNM); Museo ed Istituto di Zoologia Siste matica dell’ species distribution models based on presence data (Elith
Università, Torino (MZUT); Zoological Museum Shahid et al. 2006, Phillips et al. 2006, Elith et al. 2011). How-
Bahonar University of Kerman (ZMSBUK); Razi Universi- ever, testing is required to assess the predictive perform-
ty Zoological Museum (RUZM); Zagros Herpetological In- ance of the model. Therefore, the most usual approach is to
stitute Museum (ZHIM); and Department of the Environ- divide data into ‘training’ and ‘test’ datasets, thus creating
ment of Qom Zoological Museum (DOEQZM). A total of relatively independent data for model testing (Fielding &
33 locality records for S. microlepis were gathered and used Bell 1997, Guisan et al. 2006). Consequently, Maxent was
in the maximum entropy distribution modelling approach used with default settings when separating records into
(Maxent). 20 environmental variables, describing tempera- training and test samples (75 and 25%, respectively) with
ture, precipitation, seasonality, altitude, all with 30-arc-sec- ten replicates, which is a technique that has been proven
onds resolution, were obtained from the Worldclim data set to achieve high predictive accuracy (Phillips & Dudík
(http://www.worldclim.org/; Hijmans et al. 2005). In addi- 2008). Convergence threshold and maximum number of
tion, a slope layer was built from altitude layer information iterations were carried out by default (0.00001 and 500, re-
in ArcGIS 10 using the spatial analyst toolbox. First, corre- spectively). We used cross-validation to evaluate the pre-
lations between all 21 environmental variables were meas- dictive performance of the model. Jackknife testing was
ured with Pearson’s correlation coefficient in SPSS 16. The used to produce estimates of the average contribution and
variables with a correlation coefficient > 0.75 were excluded response of each variable to the model.
Our model was tested with ‘area’ under the receiver-
operating characteristic curve (AUC) that has been used
extensively in assaying species’ distribution models, and
measures the ability of a model to differentiate between
sites where a species is ‘present’ versus ‘absent’ (Phillips
et al. 2006, Elith et al. 2006). Models with AUC = 0.5 in-
dicate a performance equivalent to random; AUC > 0.7 in-
dicates useful performance, AUC > 0.8 indicates good per-
formance, and AUC ≥ 0.9 indicates excellent performance
(Manel et al. 2001).
The variables that contribute the most include: bio18
(40%), bio12 (20%), bio8 (15.6%), slope (10.5%), bio 17
(8.5%), bio2 (3.1%), bio14 (1.1%), bio5 (0.6%), bio7 (0.5%),
and bio15 (< 0.1%) (Table 1). The AUC value of our model
was 0.986 ± 0.005.
Modelling of the potential distribution of S. microlepis
Figure 1. Male specimen of Spalerosophis microlepis from central reveals the most suitable habitat to lie in mountainous re-
Iran, Qom. gions, including the Zagros highland and northern and
295
Correspondence
southern Afghanistan, which corresponds to the Hindu term niche conservatism is used to describe the tendency
Kush Mountains and northern Syria (Fig. 2). The doubt- of species to have similar ecological needs over evolution-
ful records are not congruent with habitat suitable for S. ary time-scales (Peterson et al. 1999, Wiens & Graham
micro lepis (Fig. 2). The environmental variables with the 2005). According to Acevedo et al. (2014), ecological data
highest gains for S. microlepis are bio18, bio17, bio 12, and suggests that niche conservatism may be explained by the
bio8 (Fig. 3) as they are those that will decrease the model’s fragmentation in the distribution range of a species’ ances-
gain the most when they are omitted; this means that these tor, which may have been the propellant of the initial stages
variables have a significant amount of information that is of divergence, without changes of the environmental niche
not represented by the other variables. of the allopatric populations. On the other hand, predicted
Our results from modelling are highly compatible with suitable areas of S. microlepis in Afghanistan and Syria are
the known distribution of S. microlepis, with the exception likely not inhabited by the species due to the lack of acces-
of predicted suitability in Afghanistan and northern Syria sibility in a biogeographical sense. The suitable areas in Za-
where the species obviously is absent. However, the close- gros Mountains are not connected by suitable habitat to the
ly related species S. diadema, which probably has similar highlands in Afghanistan and Syria. Therefore, the species
ecological traits and habitat preferences, occurs there. The could not colonize these areas.
Figure 2. Potential distribution of Spalerosophis microlepis resulting from the best-fitting Maxent model. Predicted occurrence from
low likelihood (white, 0.0) through green, orange to red (1.0). The question marks refer to doubtful records of Spalerosophis microlepis.
Figure 3. Results of Jackknife evaluations of importance of the variables used for our Spalerosophis microlepis Maxent model.
296
Correspondence
The model obtained suggests suitability for occupation We conclude that precipitation, temperature, and slope
to be the highest along the Zagros Mountains in western play the most important roles in predicting the distribu-
Iran, where most records originate. As already mentioned, tion of S. microlepis as these factors contributed about 85%
doubtful records such as Semnan, western Yazd, and north- of the environmental factors to the full model. More field-
ern Hormozgan probably do not refer to S. micro lepis and work is needed throughout Iran and Iraq to shed more
probably are based on misidentified S. diadema. In addi- light on the remaining ambiguities of the distribution of
tion, the results of Maxent modelling did not show highly S. microlepis.
suitable habitat for S. microlepis in Iraq, but isolated popu-
lations of S. microlepis probably are located in the moun-
tainous areas of the Kurdistan region, northwestern Iraq, Acknowledgements
which are considered part of the Zagros Mountains and
known to harbour many species of reptiles and amphi- This study was carried out with official permit number 40401008
issued by Department of the Environment of Iran. It was supported
bians also present in the Iranian part of the Zagros. A re-
by the Zagros Herpetological Institute Museum (ZHIM) and De-
cent study confirms the occurrence of an isolated popu-
partment of the Environment. We are very grateful to the staff of
lation of the species in northwestern Iraq (Afrasiab &
the Department of the Environment of Qom, Alireza Najimi and
Mohamad 2014). Topographically, the Zagros Mountains Seyyed Ahmad Shafiei Darabi, Eskandar Rastegar-Pouyani,
form a barrier between the central plateau and the Meso- Meysam Mashayekhi, and Masood Farhadi Qomi for their co-
potamian lowlands and a corridor for the southward dis- operation during fieldwork. We greatly appreciate Beat Schätti
tribution of northern faunal elements (Fisher 1968). The and Notker Helfenberger for helping us to gather point locali-
ties. Finally, we are grateful to Nicolas Vidal, Richard Ethe-
mountains host the highest number of endemics on the Ira-
ridge, and Steven Clement Anderson for editing the English of
nian Plateau and also are considered part of the 20th global
our manuscript and their helpful comments.
hotspot region, the so-called Irano-Anatolian biodiversity
hotspot (Mittermeier et al. 2004, Gholamifard 2011).
Iran is home to ten endemic species of snakes and seven of
References
these occur in the Zagros Mountains: Xerotyphlops wilso­
ni, Hiero phis andreanus, Eirenis rechingeri, Spalerosophis Acevedo, P., J. Melo-Ferreira, R. Real & P. C. Alves (2014): Ev-
micro lepis, Telescopus tesselatus, Pseudocerastes urarach­ idence for niche similarities in the allopatric sister species Lepus
noides, and Eirenis (Pedio phis) punctatolineatus condoni. castroviejoi and Lepus corsicanus. – Journal of Biogeo graphy, 41:
Consequently, the Zagros Mountains play a major role in 977–986.
the separation, isolation, and speciation of the Iranian her- Afrasiab, S. R. & S. I. Mohamad (2014): New records of snakes
petofauna (e.g., Anderson 1968, Rastegar-Pouyani et al. from Iraq (Reptilia: Colubridae). – Zoology in the Middle East,
60(1): 92–94.
2010, Hosseinzadeh et al. 2014).
Amongst the ten environmental variables that were used Anderson, S. C. (1968): Zoogeographic analysis of the lizard fau-
na. – pp. 305–371 in: Fisher, W. B. (ed.): The land of Iran, Vol. 1.
in this study, the most important factors were precipita-
The Cambridge History of Iran. – Cambridge University Press,
tion in the coldest quarter (bio18) and annual precipitation
Cambridge.
(bio12), as these variables contributed 40 and 20%, respec-
Baig, K. J. & R. Masroor (2008): The snakes of the genus Spalero­
tively. On the other hand, the AUC value of the full model
sophis Jan, 1865 in Indo-Pakistan and Iran (Squamata: Serpen-
was excellent and the standard deviation (SD) of the model
tes: Colubridae). – Herpetozoa, 20(3/4): 109–115.
was also very low, which implied a good performance of
Bombi, P., L. Luiselli, M. Capula & D. Salvi (2009): Predict-
the model (Manel et al. 2001). Therefore, the results of our
ing elusiveness: potential distribution model of the Southern
modelling indicated a trend for S. microlepis of preferably smooth snake, Coronella girondica, in Italy. – Acta Herpetolog-
selecting relatively humid habitats (mountainous regions: ica, 4(1): 7–13.
Zagros Mountains, Hindu Kush Mountains). However, for Brito, J. C., E. G. Crespo & O. S. Paulo (1999): Modelling wildlife
S. microlepis, precipitation in the coldest quarter (bio18), distributions: logistic multiple regression vs. overlap analysis. –
annual precipitation (bio12), and mean temperature in the Ecography, 22: 251–260.
wettest quarter (bio8) were the three factors that were more Buckley, L. B. & W. Jetz (2010): Lizard community structure
significantly associated with its distribution. Commonly, along environmental gradients. – Journal of Animal Ecology,
environmental variables such as precipitation or temper- 79(2): 358–365.
ature are responsible for the distribution patterns exhib- Elith, J., C. H. Graham, R. P. Anderson, M. Dudík, S. Ferrier, A.
ited by many reptile species (e.g., Real et al. 1997, Brito Guisan, R. J. Hijmans, F. Huettmann, J. R. Leathwick, A. Leh-
et al. 1999, Guisan & Hofer 2003, Rodríguez et al. 2005, mann, J. Li, L. G. Lohmann, B. A. Loiselle, G. Manion, C. Mo-
ritz, M. Nakamura, Y. Nakazawa, J. Overton, A. T. Pe ter-
Hosseinzadeh et al. 2014). As has already been shown by
son, S. J. Phillips, K. S. Richardson, R. Scachetti-Pe rei ra, R.
many authors, reptiles and amphibians are ectotherms and
E. Schapire, J. Soberón, S. Williams, M. S. Wisz & N. E. Zim-
absolutely depend on ambient warmth to raise their body
mermann (2006): Novel methods improve predictions of species
temperature and then become active, so that they often distribution from occurrence data. – Ecography, 29(2): 129–151.
have limited climatic tolerances and are strongly depend-
Elith, J., S. J. Phillips, T. Hastie, M. Dudĺk, Y. E. Chee & C. J.
ent on climatic conditions (Buckley et al. 2010, Luo et al. Yates (2011): A statistical explanation of Maxent for ecologists.
2012, Hosseinzadeh et al. 2014). – Diversity and Distributions, 17(1): 43–57.
297
Correspondence
Fielding, A. H. & J. F. Bell (1997): A review of methods for the as- Peterson, A. T., S. L. Egbert, V. Sanchez-Cordero & K. P. Price
sessment of prediction errors in conservation presence/absence (2000): Geographic analysis of conservation priorities using
models. – Environmental Conservation, 24(1): 38–49. distributional modeling and complementarity: endemic birds
Firouz, E. (2005): The Complete Fauna of Iran. – I. B. Tauris, Lon- and mammals in Veracruz, Mexico. – Biological Conservation,
93: 85–94.
don & New York, 322 pp.
Peterson, A. T., J. Soberon & V. Sanchez-Cordero (1999): Con-
Fisher, W. B. (1968): The land of Iran, Vol. 1. The Cambridge his-
servatism of ecological niches in evolutionary time. – Science,
tory of Iran. – Cambridge University Press, Cambridge, 766 pp.
285: 1265–1267.
Franklin, J. (1995): Predictive vegetation mapping: geographic
Phillips, S. J. & M. Dudík (2008): Modeling of species distribu-
modeling of biospatial patterns in relation to environmental
tions with Maxent: new extensions and a comprehensive evalu-
gradients. – Progress in Physical Geography, 19(4): 474–499.
ation. – Ecography, 31(2): 161–175.
Frynta, D., J. Moravec, J. Čiháková, J. Sádlo, Z. Hodková, M.
Phillips, S. J., R. P. Anderson & R. E. Schapire (2006): Maxi-
Kaftan, P. Kodym, D. Král, V. Pitule & L. Šejna (1997): Re-
mum entropy modelling of species geographic distributions. –
sults of the Czech Biological Expedition to Iran. Part 1. Notes
Ecological Modelling, 190: 231–259.
on the distribution of amphibians and reptiles. – Acta Societatis
Zoologicae Bohemicae, 61: 3–17. Rastegar-Pouyani, E., N. Rastegar Pouyani, S. Kazemi Nou-
reini, U. Joger & M. Wink (2010): Molecular phylogeny of the
Gholamifard, A. (2011): Endemism in the reptile fauna of Iran. –
Eremias persica complex of the Iranian plateau (Reptilia: Lacer-
Iranian Journal of Animal Biosystematics, 7(1): 13–29.
tidae), based on mtDNA sequences. – Zoological Journal of the
Graham C. H., S. Ferrier, F. Huettman, C. Moritz & A. T. Linnean Society, 158(3): 641–660.
Peterson (2004): New developments in museum-based in-
Rastegar-Pouyani, N., H. G. Kami, M. Rajabzadeh, S. Shafiei
formatics and application in biodiversity analysis. – Trends in
& S. C. Anderson (2008): Annotated Checklist of Amphibians
Ecology and Evolution, 19(9): 497–503.
and Reptiles of Iran. – Iranian Journal of Animal Biosystemat-
Guisan, A. & U. Hofer (2003): Predicting reptile distributions at ics, 4(1): 43–66.
the mesoscale: relation to climate and topography. – Journal of
Real, R., J. M. Pleguezuelos & S. Fahd (1997): The distribution
Biogeography, 30(8): 1233–1243.
patterns of reptiles in the Rif region, northern Morocco. – Afri-
Guisan, A., O. Broennimann, R. Engler, M. Vust, N. G. Yoc- can Journal of Ecology, 35(4): 312–325.
coz, A. Lehmann & N. E. Zimmermann (2006): Using niche-
Rissler, L. J., R. J. Hijmans, C. H. Graham, C. Moritz & D. B.
based models to improve the sampling of rare species. – Con-
Wake (2006): Phylogeographic lineages and species compari-
servation Biology, 20(2): 501–511.
sons in conservation analysis: a case study of California her-
Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones & A. Jarvis petofauna. – American Naturalist, 167(5): 655–666.
(2005): Very high resolution interpolated climate surfaces for
Rodríguez, M. Á., J. A. Belmontes & B. A. Hawkins (2005): En-
global land areas. – International Journal of Climato lo gy, 25(15):
ergy, water and large-scale patterns of reptile and amphi bian
1965–1978.
species richness in Europe. – Acta Oecologica, 28(1): 65–70.
Hosseinzadeh, M. S., M. Aliabadian, E. Rastegar-Pouyani
Santos, X., J. C. Brito, N. Sillero, J. M. Pleguezuelos, G. A.
& N. Rastegar-Pouyani (2014): The roles of environmental
Llorente, S. Fahdd & X. Parelladae (2006): Inferring hab-
factors on reptile richness in Iran. – Amphibia-Reptilia, 35(2):
itat-suitability areas with ecological modelling techniques and
215–225.
GIS: A contribution to assess the conservation status of Vipera
Kazemi, S. M., E. Rastegar-Pouyani, S. A. Shafiei Darabi, M. latastei. – Biological Conservation, 130(3): 416–425.
Ebrahim Tehrani, M. S. Hosseinzadeh, A. Mobaraki & M.
Schätti, B., F. Tillack & N. Helfenberger (2009): A contribu-
Mashayekhi (2015): Annotated checklist of amphibians and
tion to Spalerosophis microlepis Jan, 1865, with a short review of
reptiles of Qom Province, central. – Iranian Journal of Animal
the genus and a key to the species (Squamata: Serpentes: Colu-
Biosystematics, 11(1): 23–31.
bridae). – Herpetozoa, 22(3/4): 115–135.
Latifi, M. (2000): Snakes of Iran, 3rd Edition. – Department of the
Schätti, B., I. Ineich & C. Kucharzewski (2010): Nominal taxa
Environment, Tehran, 478 pp. [in Persian]
of Spalerosophis diadema (Schlegel, 1837) from Iraq to Pakistan.
Luo Z., S. Tang, C. Li, H. Fang & H. Hu (2012): Environmental ef- – Two centuries of confusion (Reptilia: Squamata: Colubrinae).
fects on vertebrate species richness: testing the energy, environ- – Revue suisse de Zoologie, 117(4): 637–664.
mental stability and habitat heterogeneity hypotheses. – Plos
Seigel, R. A. (1993): Summary: future research on snakes, or how to
One, 7(4): e35514.
combat lizard envy. – pp. 395–402 in: Seigel, R. A. & J. T. Collins
Manel, S., H. C. William & S. J. Ormerod (2001): Evaluating (eds): Snakes Ecology and Behavior. – McGraw Hill, New York.
presence-absence models in ecology: the need to account for
Sindaco, R., A. Venchi & C. Grieco (2013): The Reptiles of the
prevalence. – Journal of Applied Ecology, 38(5): 291–931.
Western Palearctic, Volume 2: Annotated Checklist and Distri-
Marx, H. (1959): Review of the colubrid snake genus Spalero­ butional Atlas of the Snakes of Europe, North Africa, Middle
sophis. – Fieldiana Zoology, 39: 347–361. East and Central Asia, with an Update to Volume 1. Monografie
Mittermeier R. A., P. Robles-Gil, M. Hoffmann, J. Pilgrim, della Societas Herpetologica Italica. II. – Edizioni Belvedere,
T. Brooks, C. G. Mittermeier, J. Lamoreux & G. A. B. Da Latina, 543 pp.
Fonseca (2004): Hotspots Revisited: Earth’s Biologically Rich- Uetz, P. (2015): The Reptile Database. Zoological Museum, Ham-
est and Most Endangered Ecoregions. – CEMEX Mexico City, burg. – Available at http://www. reptile-database.org/, last ac-
390 pp. cessed on 2 March 2015.
Moradi, N., S. Shafiei & M. E. Sehhatisabet (2013): The snake Wiens, J. J. & C. H. Graham (2005): Niche conservatism: integrat-
fauna of Khabr National Park, southeast of Iran. – Iranian Jour- ing evolution, ecology, and conservation biology. – Annual Re-
nal of Animal Biosystematics, 9(1): 41–55. view of Ecology, Evolution, and Systematics, 36: 519–539.
298